Re: ERROR: cannot insert duplicate... on VACUUM ANALYZE - Mailing list pgsql-general

From Jon Lapham
Subject Re: ERROR: cannot insert duplicate... on VACUUM ANALYZE
Date
Msg-id 3BC33D5E.1090906@extracta.com.br
Whole thread Raw
In response to ERROR: cannot insert duplicate... on VACUUM ANALYZE  (Jon Lapham <lapham@extracta.com.br>)
List pgsql-general
Tom Lane wrote:

>
>>main_v0_8=# REINDEX INDEX admin_users_pkey;
>>REINDEX
>>main_v0_8=# VACUUM ANALYZE ;
>>ERROR:  No one parent tuple was found
>>
>
> This still looks like corrupted data, but one can't tell from this which
> table it's in.  Try VACUUM VERBOSE.
>


This database is huge, so I won't fill the archives with the entire
output of VACUUM VERBOSE, but below are a few snippets.

This is an example of what most of the "non admin_users" indexes look
like, notice the "Deleted 0":

NOTICE:  Index plate_map_pkey: Pages 2; Tuples 18: Deleted 0. CPU

0.00s/0.00u sec.

These are the two "admin_users" indexes, notice the "Deleted 141":
NOTICE:  Index admin_users_pkey: Pages 2; Tuples 29: Deleted 141. CPU
0.00s/0.00u sec.
NOTICE:  Index admin_users_name_key: Pages 2; Tuples 29: Deleted 141.
CPU 0.00s/0.00u sec.
ERROR:  No one parent tuple was found

These are the last 3 messages that appear when running the VACUUM
VERBOSE, the 2 NOTICE's and the final ERROR.

--

-**-*-*---*-*---*-*---*-----*-*-----*---*-*---*-----*-----*-*-----*---
  Jon Lapham
  Extracta Moléculas Naturais, Rio de Janeiro, Brasil
  email: lapham@extracta.com.br      web: http://www.extracta.com.br/
***-*--*----*-------*------------*--------------------*---------------


pgsql-general by date:

Previous
From: "Command Prompt, Inc."
Date:
Subject: Abort State sensitivity?
Next
From: Eric D Nielsen
Date:
Subject: Problems installing with --with-python